https://twitter.com/bascule/status/1234493080583143424?s=20
Bitcoin energy consumption hits a new all time high of nearly 9GW, comparable to Chile, a country with 18M people. Carbon footprint is ~37 Mt CO2 annually, about that of New Zealand. And yet it still does ~4 transactions per second...
My longer comment on this point: https://news.ycombinator.com/item?id=19193938
Was going to reply "name three", but I see you covered that in your other post - well, two out of three; "Ferraris for show-off producers in LA, or Hello Kitty backpacks".
But those other things don't have a superlinear growth in energy consumption baked into the fundamentals of operation. A Hello Kitty backpack doesn't need schoolchildren to keep burning electricity just to secure its contents.
Most things are O(n) - O(n log n) in energy use to general utility provided, and top out at some point. Proof of Work chains need that just to keep the network working, regardless of any utility provided on top (which arguably is near-zero for any legitimate use case).
But I agree with the general point - pricing in full externalities of energy use would go a long way towards fixing things, including Bitcoin's existence.
Unrestricted social signaling games absolutely have runaway costs and zero net social benefits; if we're going to address those, we're right back to "charge for externalities", which was the solution anyway, and completely unspecific to Bitcoin.
Edit:
>But I agree with the general point - pricing in full externalities of energy use would go a long way towards fixing things, including Bitcoin's existence.
Why do you say that? it would just change the total equilibrium expenditure on mining, not render it pointless, since the blockchain just has too be too expensive to attack, and the same constraints would apply to attackers and miners.
I feel it would significantly slow its growth, perhaps making it not worth the while relative to alternative solutions.
Also, the way I understand it, PoW has its growth limited only by a) how fast can you provision the hardware, and b) how much energy you can throw at it. I worry that we'll never arrive at the point of having clean energy too cheap to meter, if we have a black hole fueled by pure, refined greed, which can suck all the energy surplus pretty much instantly.
If energy becomes more expensive -- say, by pricing in the externalities of some energy sources -- that decreases the equilibrium expenditure, because marginal mining can't pay for itself anymore.
The concept of "pricing out miners" doesn't even make sense to begin with. Miners are competing with other miners (including malicious ones). Any resource constraint affects all of them; it can't make all of them give up. To the extent that other coins become popular, that (again) just prices out marginal miners.
Mining is pretty centralised, and most players keep secret exactly how efficient their hardware is - and you can bet it'll end up much more efficient than publically available hardware.